1. Field of the Invention (Technical Field)
The invention relates to microwave generation and more particularly to a resonant frequency (RF) generator that operates at low impedance, amplifies the current to increase the RF output power, allows tuning the frequency of the apparatus, and a method to allow operation as an amplifier
2. Background Art
The efficient generation of microwaves from modulated electron beams requires electron beam velocity spectrums with low ratios of perpendicular energy to axial energy. Devices which violate this criteria pay a large price in terms of efficiency. For example, the virtual cathode oscillator (D. J. Sullivan, “High Power Microwave Generation using a Relativistic Electron Beam in a Waveguide Tube,” U.S. Pat. No. 4,345,220, 17 Aug. 1982) has a very high ratio of E-perpendicular/E-parallel at the nominal axial location of the virtual cathode, potentially exceeding unity. Due to challenges in extracting usable RF power from such beams the practical efficiency of this device, a few percent typically, is poor, and no efficient means of harnessing the high modulated currents, often exceeding a few 10s kA at voltages of order 500 kV, has been developed.
A highly efficient device for modulating electron beams is known as the Split Cavity Oscillator, as described in U.S. Pat. No. 5,235,248. While this device has a high ratio of E-perpendicular/E-parallel at its exit port, this ratio is substantially reduced with acceleration, of the modulated electron beam to voltages of order MV. Post-acceleration of a spatially modulated electron beam, as a means to lock in a spatial modulation while substantially increasing axial kinetic energy and thus reducing E-perpendicular/E-parallel, has been used for many years. As far back as 1940 Haeff and Nergaard described post-acceleration in their Inductive Output Amplifier device, as shown in “A wide-band inductive-output amplifier,” A. V. Haeff and L. S. Nergaard, Proc. of the IRE, vol. 28, pp. 126–130, March 1940. With post-acceleration, the SCO modulated beam kinetic energy can be converted to RF electromagnetic fields quite efficiently, exceeding 50%. However, virtual cathode formation limits the attainable current, due to space charge limitations in the modulating cavity of the device.
The operation of the prior art transit time oscillator (TTO), split cavity oscillator (SCO), and post accelerated split cavity oscillator (PASCO) are next briefly described in order to enable a distinction between previous techniques and the new methods described in the present invention.
The geometry of the TTO microwave oscillator is depicted in FIG. 1. Its operation relies on the interaction of a direct current (DC) electron beam 10 and the field of a cavity formed by a cylindrical pill box with perfectly conducting walls. The DC electron beam is often produced by a thermionic or field emission cathode 12. The geometry of the cavity is such that the time of flight 18 across the cavity 20 and the interaction of the beam 10 with the oscillating axial electric field associated with the cavity's axially symmetric mode 22 produce a spatially modulated electron beam 14. The spatially modulated electron beam 14 is converted to an electromagnetic wave 16; this conversion process is depicted symbolically in FIG. 1 since the details of the extraction/conversion process vary depending on the device and/or application. The operation of the TTO is described in detail in “The Split Cavity Oscillator: a high power e-beam modulator and microwave source,” B. Marder, et al., pg. 312, IEEE Trans. Plasma Sci., vol. 20, 1992. This device is an extremely inefficient oscillator, a problem that is addressed by the SCO.
The geometry of the SCO microwave oscillator is depicted in FIG. 2. Its operation also relies on the interaction of a direct current (DC) electron beam 40 and the field of a cavity 41. In this case the cavity 42 is formed by a cylindrical pill box cavity with an intermediate electrically conducting septum 44 (placed exactly midway across the pill box) that extends part ways across the interior of the cavity 42, as shown in FIG. 2. Distinct from the operation of the TTO, the SCO operates according to a well known energy instability that exists between the electron beam 40 and the cavity 42, and an externally applied field is not required, as described in “The Split Cavity Oscillator: a high power e-beam modulator and microwave source,” B. Marder, et al., pg. 312, IEEE Trans. Plasma Sci., vol. 20, 1992. Again, the DC electron beam 40 is often produced by a highly inefficient thermionic or field emission cathode 46. As with the TTO, the geometry of the SCO cavity is such that the time of flight across the cavity 48 and the interaction of the beam with the Pi-mode of the cavity's oscillating electric field cavity produce a spatially modulated electron beam 50. The SCO geometry and resulting electromagnetic field mode structure permits the axial length 48 of the cavity to be much shorter than the axial length of the TTO oscillator 18. Finally, the spatially modulated electron beam 50 is converted to an electromagnetic wave 52 as depicted symbolically in FIG. 2. Note that cathode 46 must produce the entire charge of electron beam 40, and consequently, operation at high powers requires that cathode 46 be capable of producing high current densities at high voltages. This often results in short cathode lifetimes, and pulse shortening due to gap closure caused by plasma drift across the anode-cathode gap, as shown in “Results of research on overcoming pulse shortening of GW class HPM sources,” K. Hendricks, et al., pg. 81, Digest of Technical Papers, International Workshop on High Power Microwave Generation and Pulse Shortening, Edinburgh, Scotland, 1997. The main disadvantage of the SCO is the large axial velocity spread, and the substantial unwanted perpendicular velocities of the electrons that exit the cavity, leading to poor beam-to-RF power conversion efficiencies. The SCO is described in U.S. Pat. No. 5,235,248. However, this prior art patent describes an apparatus with a very specific geometry (cylindrical pill box with an electrically conducting septum placed exactly midway along the axial length of the cavity) that operates at a single frequency. No capability to adjust the frequency of operation of the device is disclosed or implied in the prior art patent.
The geometry of the PASCO microwave oscillator is depicted in FIG. 3. Its operation to produce a spatially modulated electron beam 60 is equivalent to the SCO described above. However, once the spatially modulated electron beam 60 is produced, the PASCO uses an accelerating screen or grid 62 at a high relative potential, typically 100's of kV or more, to accelerate the electron beam to relativistic velocities 64, i.e., close to the speed of light. This greatly reduces the relative axial velocity spread intrinsic to the SCO and represents a major improvement for potential high power and high efficiency operation. The relativistic spatially modulated electron beam 64 is then converted to an electromagnetic wave 66 as depicted in FIG. 3. This technique results in a more tightly velocity-modulated beam, while maintaining excellent spatial bunching allowing more efficient beam-to-RF extraction. The main disadvantages of the PASCO are: (1) the device has an inherent limitation on total current due to space charge depression in the modulating cavity, which ultimately can lead to virtual cathode formation, but at more modest current levels, reduces modulation efficiency; and (2) the PASCO is a fixed frequency device, in that there is no ability to tune its frequency while maintaining axisymmetry.
Post-acceleration of an electron beam for high power and high efficiency operation is described in U.S. Pat. No. 5,101,168. However, this patent describes methods that were well known prior to the patent's application date. As an example, post-acceleration of an electron beam was described by Haeff and Nergaard, “A wide-band inductive-output amplifier,” A. V. Haeff and L. S. Nergaard, Proc. of the IRE, vol. 28, pp. 126–130, March 1940. Furthermore, post-acceleration of an electron beam was described by Preist and Shrader, “The Klystrode—an unusual transmitting tube with potential for UHF,” D. H. Preist and M. B. Shrader, Proc. of the IEEE, vol. 70, no. 11, pp. 1318–1325, November 1982.
The present invention, a Current Amplified, Tunable, Post Accelerated, Modulator (CATPAM) apparatus uses techniques of the well known transit time oscillator (TTO) as described in “Interchange of energy between an electron beam and an oscillating electric field,” J. Marcum, Journal of Applied Physics, vol. 17, January, 1946, a split cavity oscillator (SCO) shown in ‘The Split Cavity Oscillator: a high power e-beam modulator and microwave source,” B. Marder, et al., pg. 312, IEEE Trans. Plasma Sci., vol. 20, 1992, and the post accelerated split cavity oscillator (PASCO) (the PASCO is also known as the Reltron described in “Super RELTRON theory and experiments,” R. Miller, et al., pg. 332, IEEE Trans. Plasma Sci., vol. 20, 1992, in conjunction with unique techniques to operate at low impedance, amplify the current to increase the RF output power, tune the frequency of the device, and a method to allow operation as an amplifier, as opposed to just an oscillator. The disclosed apparatus spatially modulates a direct current (DC) electron beam using instabilities associated with device geometry and transit time effects; or, it directly generates a spatially modulated electron beam using laser-induced electron emission. It then amplifies the resulting electron beam (current), accelerates the spatially modulated beam to relativistic velocities, and converts the kinetic energy of the spatially modulated relativistic electron beam to electromagnetic fields at microwave frequencies. In addition, methods are disclosed that allow the device to be tuned to a desired operating frequency while maintaining nominal axisymmetry. None of the prior art teaches or implies these novel features.